Project Summary. Silencing of tumor suppressor genes by epigenetic deregulation is a common occurrence in human malignancies. We find that loss of CTCF-dependent chromatin boundaries, which protects genes from adjacent heterochromatin domains, results in transcriptional inactivation of the p16INK4a tumor suppressor gene, which is a frequent target of epigenetic silencing in many types of human cancers and considered to be an early event in breast carcinogenesis. Loss of CTCF binding also correlates with hypermethylation and silencing of two other tumor suppressor genes, RASSF1A and CDH1, in breast cancer cell lines. CTCF dissociation from the boundary elements of these tumor suppressor genes results from defective poly(ADP-ribosyl)ation of CTCF, which abrogates its proper function. In this proposal, we plan to use primary human mammary epithelial cells (HMECs) to analyze the role of CTCF in molecular aberrations that initiate breast tumorigenesis. In this system, distinct subpopulations of cells emerge from normal HMECs that have overcome barriers to indefinite growth. The first barrier exhibited by variant HMECs coincides with p16 gene silencing followed by increasing epigenetic plasticity and chromosomal aberrations similar to those observed in early human breast cancers. Our specific aims are to: characterize native CTCF protein complexes from HMECs and p16-silenced vHMECs using biochemical approaches (Aim 1); examine the basis of defective CTCF PARlation in vHMECs using cell- based (Aim 2) [and biochemical studies (Aim 3)]; and [analyze the role of CTCF protein complexes in p16 gene regulation (Aim 4).] We postulate that destabilization of specific chromosomal boundaries by dysfunctional CTCF complexes may be an initiating event in the genesis of human breast cancers by early inactivation of critical tumor suppressor genes and loss of normal genomic patterns of epigenetic regulation.